A large-scale, higher-level, molecular phylogenetic study of the insect order Lepidoptera (moths and butterflies)

Abstract

Background: Higher-level relationships within the Lepidoptera, and particularly within the species-rich subclade Ditrysia, are generally not well understood, although recent studies have yielded progress. We present the most comprehensive molecular analysis of lepidopteran phylogeny to date, focusing on relationships among superfamilies.

Methodology principal findings: 483 taxa spanning 115 of 124 families were sampled for 19 protein-coding nuclear genes, from which maximum likelihood tree estimates and bootstrap percentages were obtained using GARLI. Assessment of heuristic search effectiveness showed that better trees and higher bootstrap percentages probably remain to be discovered even after 1000 or more search replicates, but further search proved impractical even with grid computing. Other analyses explored the effects of sampling nonsynonymous change only versus partitioned and unpartitioned total nucleotide change; deletion of rogue taxa; and compositional heterogeneity. Relationships among the non-ditrysian lineages previously inferred from morphology were largely confirmed, plus some new ones, with strong support. Robust support was also found for divergences among non-apoditrysian lineages of Ditrysia, but only rarely so within Apoditrysia. Paraphyly for Tineoidea is strongly supported by analysis of nonsynonymous-only signal; conflicting, strong support for tineoid monophyly when synonymous signal was added back is shown to result from compositional heterogeneity.

Conclusions significance: Support for among-superfamily relationships outside the Apoditrysia is now generally strong. Comparable support is mostly lacking within Apoditrysia, but dramatically increased bootstrap percentages for some nodes after rogue taxon removal, and concordance with other evidence, strongly suggest that our picture of apoditrysian phylogeny is approximately correct. This study highlights the challenge of finding optimal topologies when analyzing hundreds of taxa. It also shows that some nodes get strong support only when analysis is restricted to nonsynonymous change, while total change is necessary for strong support of others. Thus, multiple types of analyses will be necessary to fully resolve lepidopteran phylogeny.

Figure 1. Previous hypotheses of deep-level relationships in Lepidoptera.

A. Composite working hypothesis based on morphology . B. Ditrysian-only relationships (rooted on Tineoidea) inferred from degen1 ML analysis of 123 taxa sequenced either for 5 or 26 gene segments, with bootstrap values ≥50% displayed for nodes at the superfamily level and above . C. Lepidopteran relationships (rooted on Micropterigoidea) inferred from ML analysis of 350 taxa, using nucleotides from the first and second codon positions (+ third codon position for EF-1α only) of 8 gene segments, with bootstrap values >50% displayed for nodes at the superfamily level and above . Numbers in parentheses after taxon names are numbers of exemplars sampled.

Figure 2. Assessing the effectiveness of the GARLI heuristic ML search through an analysis of 4608 search replicates as derived from the full 483-taxon, 19-gene, nt123_degen1 data set.

Out of 4608 search replicates, the single fully-resolved topology of highest likelihood is displayed (lnL  =  −583,900.053394). Terminal taxa, not shown in this figure in order to save space, are displayed in Figure S1. Dichotomous nodes that are not present in one or more strict consensuses of subsets of the 4608 topologies are identified by having numbers with blue coloration above subtending branches. The three subsets are as follows: 4, all topologies with lnL scores that are within 0.0001% (10^-4 %) of that of the best ML topology (2 topologies total, including the best ML topology); 3, all topologies within 0.001% (10^-3 %, 19 topologies total); 2, all topologies within 0.01% (10^-2 %, 1827 topologies total). Selected bootstrap percentages based on 15 heuristic search replicates and 500 bootstrap pseudoreplicates that are ≥50% are displayed below branches (see Figure S1 for all bootstrap percentages). An orange-colored bar is placed beside each node that has bootstrap support ≥50% and that is missing in one or more of the subset consensuses. The dashed arrow identifies the altered placement of one (and only one) taxon that was found in a new and improved topology (lnL  =  −583,898.838616), when the dichotomous topology displayed in this figure was used as a starting tree in a second round of 561 GARLI ML heuristic searches. This new topology was recovered in 248 of the 561 search replicates. Higher-level taxon names, some of which are abbreviated, are displayed. All abbreviations follow: Anth+Phidit+Carth+Endrom., Anthelidae + Phiditidae + Carthaeidae + Endromidae; Saturn+Sphing+Bomb., Saturniidae + Sphingidae + Bombycidae; Eupterot+Brahm+Apat., Eupterotidae + Brahmaeidae + Apatelodidae; Nolidae+Stictopt., Nolidae + Stictopterinae; Notodont+Oenosand., Notodontidae + Oenosandridae; Uraniid.+Epiplemidae, Uraniidae + Epiplemidae; Sematur.+Epicopeiidae, Sematuridae + Epicopeiidae; Papilion., Papilionidae; Pterophorid+Copromorph., Pterophoridae (part) + Copromorphidae (part); Copromorph., Copromorphidae (part); Callidul., Callidulidae; “small ditrysian families”, Copromorphidae + Carposinidae + Epermeniidae + Alucitidae + Hyblaeidae + Pterophoridae (part) + Thyrididae + Pseudurgis (unplaced); Dalcer.+Limacodidae, Dalceridae + Limacodidae; Megalopyg+Aidos+Himant., Megalopygidae + Aidos + Himantopteridae; Zygaen.+Lacturidae, Zygaenidae + Lacturidae; Choreutid+Schreckenstein.+Douglasiidae, Choreutidae + Schreckensteiniidae + Douglasiidae.

Figure 3. Summary of three phylogenetic analyses of 483 taxa and 19 genes.

Bootstrap percentages derived from GARLI analysis of three data sets -- nt123_degen1, nt123, and nt123_partition -- are displayed in that order above internal branches of a condensed, higher-level-only portion of the nt123_degen1 ML topology (see numbers in black). Selected nodes are arbitrarily numbered for convenient reference (see numbers in blue). The full nt123_degen1 and nt123 topologies are shown in Figure S1 and S2, respectively. A bracket indicates that the node displayed was not recovered in the ML analysis of that data set. A dash indicates that the bootstrap value is <50%. The number of exemplars is listed in parentheses after the family or subfamily name. The region of the topology that includes Tineoidea has blue-colored branches, and its favored alternative topology, based on analysis of nt123, is also displayed (see lower boxed area). Throughout this report, we have subsumed all tineoid taxa traditionally identified as Acrolophidae under Tineidae, all tineoid taxa traditionally identified as Arrhenophanidae under Psychidae, and Crinopterygidae under Incurvariidae, following van Nieukerken et al. . BP, bootstrap percentage.

Figure 4. Base-composition distance diagrams derived from analysis of the nt123 and nt123_degen1 data sets for 483 taxa.

Branching structure obtained by neighbor-join / minimum evolution analysis of Euclidean distances calculated on the proportions of each of the four nucleotide types in each species. All diagrams are drawn to the same scale, and units are 'per cent ÷ 100'. The blue shaded portions identify taxa deleted from nt 123 data subsets to explore the effect of decreased nucleotide heterogeneity on bootstrap percentages.

Figure 5. Summary of phylogenetic analyses based on taxon (sub)sampling of Tineoidea.

Summary phylogenetic trees are displayed with corresponding bootstrap percentages for analysis of nt123 and nt123_degen1 data sets based on different taxon subsamples for Tineoidea. For ease and focus of presentation, only relationships among strongly supported, higher-level groupings are shown (see Figure 3). These groupings are: Tine: Tineidae – Eudarcia (20 taxa total); Psych  =  Psychidae (9 taxa total); Eudarcia (currently classified within Tineidae, 1 taxon); Compsoctena (currently classified within Eriocottidae; 1 taxon); NTD: non-tineoid Ditrysia (27 taxa total); and the non-ditrysian outgroup (not shown), which consist of Palaephatidae + Tischeriidae (5 taxa total).

Figure 6. Base-composition distance diagrams of nt123_degen1 and nt123 data sets for the 63 taxa in the Tineoidea test set.

Both diagrams are drawn to the same scale, and units are 'per cent ÷ 100'. Bootstrap percentages ≥50% are displayed. Bootstrap percentages are based on analysis of total taxon-specific nucleotide compositions, as described in Materials and Methods. All terminal taxa are identified to genus for nt123 but not for nt123_degen1, due to the reduced compositional heterogeneity in the latter data set. The vertical bars identify those taxa used in a phylogenetic analysis (Figure 5) to test the effect of reduced compositional heterogeneity on the analysis of nt123. The five sets of taxa whose inter-relationships are analyzed in Figure 5 are color- and/or symbol-coded (see key).